zichen-gan opened a new pull request, #2165: URL: https://github.com/apache/zookeeper/pull/2165
Issue: - The current ZooKeeper (ZK) cluster frequently experiences excessively long Full Garbage Collection (FGC) events (averaging once a week) that last over 10 seconds, leading to occasional, unexpected leader elections in the ZK cluster. - The cluster currently fails to collect mntr monitoring data. Root Cause: - The ZK cluster uses the direct child nodes under the root directory (/) as Namespaces, At the Namespace level, the following have been implemented: - Monitoring of MIN, CNT, AVG, MAX, and SUM request counts, which can be obtained via mntr monitoring. - monitoring for Namespaces only accounts for additions and not removals, tasks are created directly under the root directory (/), with tasks older than seven days being cleaned daily. - This leads to the following phenomenon: The number of Namespaces stored in ZK nodes (157,376) is significantly greater than the actual number of Namespaces in the ZK cluster (5,856). Solution: Provide a new version of the ZK server that supports configurable Namespace monitoring. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: notifications-unsubscr...@zookeeper.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
